TikaFlow/dsh-model-reasoning
为所有非官方（自定义）提供商的模型自动填充推理级别，数据来自 models.dev。 / Automatically fills in reasoning-effort levels for models from unofficial (custom) providers, with data sourced from models.dev.
Listed
0
Model
Bundle verified
What it does
Automatically fills in reasoning-effort levels for models from unofficial (custom) providers, with data sourced from models.dev.
Best for
- Custom-provider users whose model entries lack selectable reasoning-effort levels.
- Installations that want reasoning metadata populated automatically from models.dev data.
- Mixed configurations where existing manual `reasoningEfforts` values must remain unchanged.
Not ideal for
- Official-provider models, since the plugin targets unofficial custom providers.
- Models whose `reasoningEfforts` are already configured, because those values are intentionally left untouched.
- Models for which the bundled or refreshed models.dev catalog has no applicable reasoning data.
README
dsh-model-reasoning
DSH 插件：为所有非官方（自定义）提供商的模型自动填充推理级别（reasoningEfforts），推理级别数据来自 models.dev。
功能
- 进入插件时优先使用构建附带的 models.dev 缓存（存在、大小不低于 10KB 且解析成功才算可用）立即填充推理级别，避免启动时等待网络
- 延迟 5 秒异步拉取 models.dev 最新数据：成功则更新缓存并再次填充；失败则记录日志，继续使用现有目录
- 监听
settings/updated事件，模型配置变更后自动重新填充 - 仅填充缺少
reasoningEfforts的模型，已有配置不受影响
安装
dsh plugin --profile web add github:TikaFlow/dsh-model-reasoning
使用说明
无需任何操作，重启 DSH 后插件即自动生效：支持推理级别的模型将会自动填充推理级别，可在界面中选择。
Frequently Asked QuestionsFAQ
Use the verified command dsh plugin --profile default add github:TikaFlow/dsh-model-reasoning in a DSH-enabled shell. The command resolves the public package metadata and keeps the plugin attached to the catalog identity shown on this page.
Compatibility follows the bundle and profile status shown above. If a profile is not detected, keep the plugin disabled there and check the repository documentation before enabling it in production.
The GitHub link and activity metadata are the source of truth for releases and maintenance. Revisit this page after a new release to confirm the catalog has observed the latest version.